Switchback experiments — when and how?

hard

Answer

$Alternate treatment on/off over time windows across the whole market \to each window randomly assigned.$
$Estimator: difference in outcomes between treated and control windows, with cluster-robust or time-block SEs.$
$Uses: full-market experiments where user-level randomization fails (surge pricing at Uber, driver dispatch at DoorDash).$
$Carryover between windows biases the estimate — use burn-in periods within each window and appropriate window length.$

Check yourself — multiple choice

Same as A/B
Alternate treatment on/off over time in whole market; compare treated vs control windows with block-robust SEs; used for surge pricing / dispatch experiments
Random
Not real

Switchback: alternate on/off over time windows; used in marketplace experiments.

#ab-testing#quasi-experimental

215 interview questions in this topic.